Impact statement
South American savannas are ecologically rich and functionally diverse ecosystems, yet they are being increasingly threatened by the spread of invasive perennial grasses. These invaders drive profound ecological transformations by altering fire regimes, accelerating nutrient cycling and disrupting native vegetation dynamics, ultimately leading to biodiversity loss, soil degradation and the decline of key ecosystem services. We synthesize current research on how invasive perennial grasses reshape ecological processes and interact with climate change and land-use transformation. We also address the challenges faced to restore invaded areas. Effective responses to grass invasions are urgently needed to sustain restoration efforts and safeguard ecosystem services in these dryland regions. Strengthening our ecological understanding of invasion dynamics is essential to guide conservation policy and promote sustainable land management in South America.
Introduction
Biological invasions have intensified over recent decades, emerging as one of the leading drivers of biodiversity loss and disruption of ecological processes in natural ecosystems (Pyšek et al., Reference Pyšek, Hulme, Simberloff, Bacher, Blackburn, Carlton, Dawson, Essl, Foxcroft, Genovesi, Jeschke, Kühn, Liebhold, Mandrak, Meyerson, Pauchard, Pergl, Roy, Seebens, van Kleunen, Vilà, Wingfield and Richardson2020; Daly et al., Reference Daly, Chabrerie, Massol, Facon, Hess, Tasiemski, Grandjean, Chauvat, Viard, Forey, Folcher, Buisson, Boivin, Baltora-Rosset, Ulmer, Gibert, Thiébaut, Pantel, Heger, Richardson and Renault2023; IPBES Reference Roy, Pauchard, Stoett and Renard2023). Among the most significant groups, invasive perennial grasses (Poaceae) stand out due to their high dispersal capacity, disturbance tolerance and strong competitive ability, which are traits that facilitate their establishment and spread, particularly in degraded areas and fire-prone open ecosystems such as savannas (D’Antonio and Vitousek, Reference D’Antonio and Vitousek1992; Williams and Baruch, Reference Williams and Baruch2000; Sampaio-Franco et al., Reference Sampaio-Franco, Pivello, Magalhães, Castro, Cruz Neto, Matos, Brown, Heringer, Saulino, Oliveira, Braga, Miranda, Mormul and Vitule2024).
In South American savannas, negative impacts of invasive grasses are particularly concerning, as these ecosystems face intense anthropogenic pressure from expanding livestock farming and the alteration of natural fire dynamics (Franco et al., Reference Franco, Rossatto, Silva and Ferreira2014; Zenni et al., Reference Zenni, Herrera, Dechoum, Ziller, ACL, Núñez, Pauchard, Clements, Upadhyaya, Joshi and Shrestha2022). The impacts are often exacerbated by the absence of robust, well-informed and targeted conservation strategies and public policies that directly tackle the underlying causes of degradation of open ecosystems and their recovery (Williams et al., Reference Williams, Watson, Beyer, Grantham, Simmonds, Alvarez, Venter, Strassburg and Runting2022; Overbeck and Pillar, Reference Overbeck and Pillar2024).
When introduced into new environments, such as savanna regions in South America, these grasses may shift from being merely exotic (not native) to becoming invasive, acting as drivers of profound ecological transformation. As will be further explored in the following topics, non-native perennial grasses were generally introduced in South American savannas to serve as forage for livestock. However, they frequently escape cultivation and spread into adjacent natural habitats, leading to biomass accumulation, disruption of fire regimes and acceleration of nutrient cycling. These changes establish self-reinforcing feedback loops that hinder natural regeneration, ensure the persistence of invasive populations and trigger cascading effects, including the simplification of plant and animal communities and degradation of key ecosystem services (D’Antonio and Vitousek, Reference D’Antonio and Vitousek1992; Klink and Machado, Reference Klink and Machado2005; Bond and Midgley, Reference Bond and Midgley2012; Soares et al., Reference Soares, Nascimento, da Silva and de Oliveira2021; Zenni et al., Reference Zenni, Herrera, Dechoum, Ziller, ACL, Núñez, Pauchard, Clements, Upadhyaya, Joshi and Shrestha2022). Climate change, manifested in rising atmospheric aridity and the expansion and intensification of the dry season across much of seasonally dry tropical South America (Duffy et al., Reference Duffy, Brando, Asner and Field2015; Marengo et al., Reference Marengo, Souza, Thonicke, Burton, Halladay, Betts, Alves and Soares2018, Reference Marengo, Jimenez, Espinoza, Cunha and Aragão2022; Hofmann et al., Reference Hofmann, Cardoso, Alves, Weber, Barbosa, de Toledo, Pontual, Salles, Hasenack, Cordeiro, Aquino and Oliveira2021, Reference Hofmann, Weber, Bastazini, Rossatto, Franco, Granada, Kaminski, Ubaid, Leandro-Silva, Borges-Martins, Silva, Cardoso, Oliveira, Aquino and Pereira2025), creates more challenging conditions for ecosystems, making these invasive grasses even harder to manage.
Against this backdrop, there is an urgent need to deepen our understanding of the mechanisms through which invasive perennial grasses alter the structure, composition and functioning of South American savannas and to develop effective management and restoration strategies that curb their expansion and prevent reinvasion of restored areas.
We synthesize current knowledge on the impacts of the major invasive grasses in South American savannas, with particular emphasis on the Brazilian Cerrado, home of the largest area of savanna in the Neotropics. We center our discussion on structural and functional alterations to these ecosystems, their interactions with fire regimes and the challenges posed by climate change. Additionally, we examine mitigation and restoration strategies that have been proposed or implemented, aiming to support biodiversity conservation and strengthen sustainable management practices in savanna environments.
Distribution, ecology and vulnerability of South American savannas
Ranging from 5° N to 25° S and reaching elevations of 2,000 m, South American savannas exhibit a high degree of environmental and ecological heterogeneity. They are distributed across several countries: in Brazil, the Cerrado, Amazonian and Pantanal savannas; in Venezuela, the Orinoco Llanos and the Gran Sabana; in Colombia, the Llanos Orientales; in Bolivia, the Llanos de Moxos (also known as the Beni Savannas); and in the Guianas, the Rupununi savannas, which are contiguous with the Lavrados (savannas of Roraima state, in northern Brazil) and the Gran Sabana (Baruch, Reference Baruch2005; Lloyd et al., Reference Lloyd, Goulden, Ometto, Patiño, Fyllas, Quesada, Keller, Bustamante, Gash and Silva Dias2009; Junk et al., Reference Junk, Piedade, Lourival, Wittmann, Kandus, Lacerda, Bozelli, Esteves, Nunes da Cunha, Maltchik, Schöngart, Schaeffer-Novelli and Agostinho2014; Borghetti et al., Reference Borghetti, Barbosa, Ribeiro, Ribeiro, Walter, Scogings and Sankaran2019). Enclaves of savanna are also found in the Gran Chaco, a vast interior plain spanning Bolivia, Paraguay and northern Argentina (Oyarzabal et al., Reference Oyarzabal, Clavijo, Oakley, Biganzoli, Tognetti, Barberis, Maturo, Aragón, Campanello, Prado and León2018; Borghetti et al., Reference Borghetti, Barbosa, Ribeiro, Ribeiro, Walter, Scogings and Sankaran2019).
South American savannas are characterized by uneven tree cover and a continuous cover of shade-intolerant C4 grasses often associated with a diverse assemblage of C3 herbs. Plant phenology and growth cycles are driven by a distinct seasonal climate with annual rainfall ranging from 800 to 2,200 mm. While soil types vary, acidic, high-aluminum, nutrient-poor soils prevail (Borghetti et al., Reference Borghetti, Barbosa, Ribeiro, Ribeiro, Walter, Scogings and Sankaran2019). Most species are perennial, with extensive below-ground systems that support resprouting after dry periods or fire events. Depending on the type of subterranean system, these structures may also play a crucial role in vegetative propagation and population expansion (Eiten, Reference Eiten1972; Sarmiento, Reference Sarmiento and Solbrig1984; Moraes et al., Reference Moraes, Carvalho, Franco, Pollock and Figueiredo-Ribeiro2016; Pilon et al., Reference Pilon, Cava, Hoffmann, Abreu, Fidelis and Durigan2021).
Fire, whether of natural or anthropogenic origin, plays a central role in shaping the dynamics of savanna ecosystems. It acts as a key structuring agent by limiting the encroachment of woody species and promoting the regeneration of the ground-layer vegetation (Bond and Parr, Reference Bond and Parr2010; Damasceno and Fidelis, Reference Damasceno and Fidelis2020; Zenni et al., Reference Zenni, Herrera, Dechoum, Ziller, ACL, Núñez, Pauchard, Clements, Upadhyaya, Joshi and Shrestha2022).
Despite their essential role in carbon cycling, water regulation and biodiversity maintenance, South American savannas remain undervalued and highly threatened. In the savanna-dominated Cerrado, nearly 50% of the original vegetation has been converted to anthropogenic land uses, with native vegetation experiencing a net loss of almost 30% over the past four decades (Strassburg et al., Reference Strassburg, Brooks, Feltran-Barbieri, Iribarrem, Crouzeilles, Loyola, Latawiec, Oliveira Filho, Scaramuzza, Scarano, Soares-Filho and Balmford2017; MAPBiomas, 2025), while just 8.6% of the total area is formally protected (Dos Santos and Cherem, Reference Dos Santos and Cherem2023). Even protected areas (PAs) in the Cerrado and in other savannas of South America are undergoing some level of degradation (Williams et al., Reference Williams, Watson, Beyer, Grantham, Simmonds, Alvarez, Venter, Strassburg and Runting2022), underscoring the urgent need for effective conservation strategies and sustainable land-use planning.
Amazonian savannas (Figure 1) provide examples of the growing vulnerability of South American savannas. They occupy less than 5% of the Amazon (~267,000 km2) in scattered patches throughout northern Brazil, Bolivia, Guyana, Suriname and Venezuela (Carvalho and Mustin, Reference Carvalho and Mustin2017; Carvalho et al., Reference Carvalho, Costa-Neto, Dagosta, Fearnside, Hilário, Oliveira, Silva, Toledo, Xavier and Mustin2025). Their patchy distribution, which contrasts with the prevailing image of a continuous rainforest, has hindered their inclusion in conservation agendas (Hilário et al., Reference Hilário, Toledo, Mustin, Castro, Costa-Neto, Kauano, Eilers, Vasconcelos, Mendes-Junior, Funi, Fearnside, Silva, Euler and Carvalho2017; Carvalho et al., Reference Carvalho, Costa-Neto, Dagosta, Fearnside, Hilário, Oliveira, Silva, Toledo, Xavier and Mustin2025). In Amapá, Brazil, only 9.2% of the 10,021 km2 of native savanna is legally protected, leaving it vulnerable to rapid agricultural expansion. Projections indicate that soybean plantations could cover 40% of the total savanna area by 2026, driving further habitat loss, intensified fire usage and the proliferation of invasive grasses (Hilário et al., Reference Hilário, Toledo, Mustin, Castro, Costa-Neto, Kauano, Eilers, Vasconcelos, Mendes-Junior, Funi, Fearnside, Silva, Euler and Carvalho2017; Mustin et al., Reference Mustin, Carvalho, Hilário, Costa-Neto, Silva, Vasconcelos, Castro, Eilers, Kauano, Mendes-Junior, Funi, Fearnside, Silva, Euler and Toledo2017). Similar dynamics occur in the other South American northern savannas, such as those in French Guiana and Guyana, where agricultural expansion, infrastructure development and weak governance contribute to ongoing habitat loss, with less than 3% of the savanna area in these countries under formal protection (Stier et al., Reference Stier, de Carvalho, Rostain, Claessens, Dewynter, Catzeflis, Mckey, Mustin, Palisse and de Thoisy2020; Protected Areas Trust, 2025).
Figure 1. Representative Amazonian savannas: (A) Rupununi savanna, at the border between Venezuela and Brazil. (B) Lavrados of Roraima, (C) Alter do Chão savannas and (D) Amapá savannas in northern Brazil.
Continental-scale actions integrating ecological knowledge with regional socioeconomic realities are essential to ensure the long-term persistence of South American savannas. Within this broader context, the Cerrado, owing to its vast extent and long history of ecological research, serves as a model for understanding the ecological complexity and conservation challenges of tropical savannas worldwide.
The Cerrado: The world’s most diverse tropical savanna
The Cerrado, the largest and most biodiverse tropical savanna on Earth, spans over 2 million km2 across central Brazil, harboring about 12,000 vascular plant species, with approximately 35% of them being endemic (Mendonça et al., Reference Mendonça, Felfili, Walter, Júnior, Rezende, Filgueiras, Nogueira, Fagg, Sano, Almeida and Ribeiro2008). The high floristic richness and rapid habitat loss have earned it global recognition as a biodiversity hotspot (Myers et al., Reference Myers, Mittermeier, Mittermeier, da Fonseca and Kent2000). Land use in the region is dominated by intensive conversion of natural areas for crops and cultivated pastures, which significantly contributes to environmental degradation and biological invasion (Driscoll et al., Reference Driscoll, Catford, Barney, Hulme, Inderjit, Pauchard, Pyšek, Richardson, Riley and Visser2014). The Cerrado ranks among the most susceptible ecosystems to invasion by alien perennial grasses (Klink and Machado, Reference Klink and Machado2005; Ratter et al., Reference Ratter, Bridgewater and Ribeiro2006; Pompeu et al., Reference Pompeu, Assis and Ometto2024). While Brazilian environmental legislation mandates the preservation of 80% of native vegetation on private properties within the government-defined Amazonian legal region, this requirement is reduced to 35% for Cerrado areas located inside this region, and to merely 20% for Cerrado vegetation elsewhere in the country (Native Vegetation Protection Law, No. 12,651/2012). This regulatory disparity has allowed large-scale expansion of agriculture, accelerating the degradation of the Cerrado (Machado et al., Reference Machado, Aguiar and Bustamante2024).
Livestock production in Brazil depends mainly on raising beef cattle on cultivated pastures – which corresponds to approximately 177 million hectares – of which 60% show varying degrees of degradation (Bolfe et al., Reference Bolfe, Victoria, Sano, Bayma, Massruhá and Oliveira2024). In the Cerrado, most pastures are established through the replacement of native vegetation by C₄ grasses from East and Central Africa, which have escaped from cultivated areas and become established in the margins of forests and savannas, facilitated by disturbances such as fire and suppression of native vegetation (Pivello et al., Reference Pivello, Shida and Meirelles1999; Williams and Baruch, Reference Williams and Baruch2000; Hoffmann et al., Reference Hoffmann, Lucatelli, Silva, Azevedo, Marinho, Albuquerque, Lopes and Moreira2004; Foxcroft et al., Reference Foxcroft, Richardson, Rejmánek and Pyšek2010; Rossi et al., Reference Rossi, Martins, Viana, Rodrigues and Figueira2014; Fernandes et al., Reference Fernandes, Tameirão, Costa, Ribeiro, Neves, Souza and Negreiros2021), along with the expansion of roads and highways that function as corridors for their dispersal (Hoffmann et al., Reference Hoffmann, Lucatelli, Silva, Azevedo, Marinho, Albuquerque, Lopes and Moreira2004; Musso et al., Reference Musso, de Macedo, Almeida, Rodrigues, Camargo, Pôrto and Miranda2019). Furthermore, in most cases, abandoned pasturelands do not spontaneously recover into species-rich, old-growth savanna communities. Even after decades of abandonment, these areas often remain dominated by invasive grasses (Cava et al., Reference Cava, Pilon, Ribeiro and Durigan2018; Silva et al., Reference Silva, Rodrigues, Bringel, Sampaio, Sano and Vieira2023).
African grasses have also been widely employed in soil conservation and erosion control programs. Melinis minutiflora P.Beauv., for instance, was used to stabilize slopes and road verges but subsequently colonized natural areas, where its presence increased vegetation flammability, thereby intensifying and increasing the frequency of fires (Parsons et al., Reference Parsons1972; Hoffmann et al., Reference Hoffmann, Lucatelli, Silva, Azevedo, Marinho, Albuquerque, Lopes and Moreira2004). Urochloa decumbens (Stapf) R.D.Webster has similarly been used for erosion prevention and soil rehabilitation in post-mining landscapes (Ferreira et al., Reference Ferreira, Parolin, Matos, Cunha, Chaves and Neckel2016).
Major invasive perennial grasses in South American savannas
The identification of the perennial grass species currently posing major invasion threats in South American savannas was based on the study of Lopes et al. (Reference Lopes, Demarchi, Piedade, Schöngart, Wittmann, Munhoz, Ferreira and Franco2023), which used the database of invasive species in Brazil from the Horus Institute for Development and Environmental Conservation (2024) to select ten Poaceae species with high or very high invasive potential, mapped their current distribution and modelled their potential distribution under current and future climate conditions in the Neotropical region. Their selection criteria included dispersal capacity, ecological impact and difficulty of control. The selected species were: Andropogon gayanus Kunth, Arundo donax L., Hyparrhenia rufa (Nees) Stapf, Megathyrsus maximus (Jacq.) B.K.Simon & S.W.L.Jacobs, Melinis minutiflora P.Beauv., Melinis repens (Willd.) Zizka, Urochloa brizantha (Hochst. ex A.Rich.) R.D.Webster, Urochloa decumbens (Stapf) R.D.Webster, Urochloa humidicola (Rendle) Morrone & Zuloaga and Urochloa ruziziensis (R.Germ.& Evrard) Crins. As our review focuses on perennial grasses, we chose not to include M. repens, which is an annual or short-lived perennial species.
The nine species are widely distributed in South American savannas and have climatically suitable areas that greatly overlap (Lopes et al., Reference Lopes, Demarchi, Piedade, Schöngart, Wittmann, Munhoz, Ferreira and Franco2023). While A. donax originates from Asia, the remaining eight invasive grasses are of African origin (Table 1). It was also the only species without a documented collection in Amazonian savannas (Lopes et al., Reference Lopes, Demarchi, Piedade, Schöngart, Wittmann, Munhoz, Ferreira and Franco2023), although it has been listed as a naturalized species in the alien flora of the Guiana Shield compiled by Delnatte and Meyer (Reference Delnatte and Meyer2012).
Table 1. General characterization of the main invasive perennial grasses in South American savannas
* Based on herbarium records that are listed in Supplementary Material.
** References: 1. Parsons, Reference Parsons1972; 2. Williams & Baruch, Reference Williams and Baruch2000; 3. Ferreira et al., Reference Ferreira, Moraes, Chiari, Simeão, Vigna and de Souza2021; 4. Jiménez-Ruiz et al., Reference Jiménez-Ruiz, Hardion, Del Monte, Vila and Santín-Montanyá2021; 5. Zenni et al., Reference Zenni, Herrera, Dechoum, Ziller, ACL, Núñez, Pauchard, Clements, Upadhyaya, Joshi and Shrestha2022; 6. Calazans et al., Reference Calazans, Lopes, Girotto, de Paula, Franco and Ferreira2023; 7. https://tropicalforages.info.
Searches in herbarium databases (Supplementary Material) permitted the tracking of the earliest documented records of these grasses in South American savannas and provided morphological descriptions, illustrations and nomenclatural data.
We also performed a broad search in the Scopus database (https://www.scopus.com/), accessed on 5 February 2025), using the following keyword combination: Invasive or Exotic or Alien and Savanna. We retrieved a total of 933 document results from 1973 to 2025. Although not an exhaustive literature search, it provides a representative sample of the published scientific knowledge on invasive species in savanna ecosystems. Because we were interested in discussing the current state-of-the-field on the impacts of invasive grasses in South American savannas, we focused on the most recent references (2020–2025, a total of 284 document results), which were then screened for the purposes of our review. Additional, updated literature was added while writing and revising the text. We relied primarily on studies in South American savannas, including studies in other regions only when essential for context.
These sources provided data on key functional traits of the species, including photosynthetic pathway (C₃ or C₄), the most important features that enhance survival or reproductive success, reproductive systems (Table 1) and identification of recent research trends regarding their ecological impacts, management challenges and the influence of climate change on the intensification of invasions.
History of invasive grass introduction into South American savannas
Although exact historical records are lacking, African grasses were accidentally brought to the Americas in the 17th century when they were conveyed with seeds for farming or as bedding aboard slave ships (Parsons, Reference Parsons1972). From the 19th century onwards, however, introductions became deliberate, primarily aimed at enhancing the productivity of livestock systems (Parsons, Reference Parsons1972; Baker, Reference Baker and Wilson1978; Table 1).
Many African grasses were introduced as forage, due their high biomass production and tolerance to abiotic stress, but they often failed to meet agronomic expectations in South American savannas. The combination of highly acidic, nutrient-poor soils, prolonged dry seasons and continuous grazing pressure constrained their persistence and reduced forage quality (Williams and Baruch, Reference Williams and Baruch2000; Baruch, Reference Baruch2005; Lee Reference Lee2018). Over time, several of these species escaped cultivation, colonized adjacent natural habitats and established self-sustaining populations (Baruch, Reference Baruch, Solbrig, Medina and Silva1996; Zenni et al., Reference Zenni, Herrera, Dechoum, Ziller, ACL, Núñez, Pauchard, Clements, Upadhyaya, Joshi and Shrestha2022). Herbarium records document this trajectory (Supplementary Material). Of the nine species evaluated in this review, H. rufa, M. minutiflora and M. maximus were among the first African grasses to be documented in South American savannas, with collections dating to the early 19th century. A. gayanus was introduced later, in the 1940s–1950s, with occurrences reported in Venezuela, Uruguay and Brazil. Records of this species were documented in 1968 in the Cerrado (Table 1; Supplementary Material).
Herbarium records indicate that U. brizantha and U. decumbens were present in Brazil as early as the first half of the 20th century, followed by U. humidicola and U. ruziziensis in the 1960s (Supplementary Material). By the end of the decade, all four species had been recorded in the Cerrado, and have since become dominant in the seed production and livestock sectors, due to their low cost, adaptability to low-fertility soils and high efficiency in ruminant nutrition (Jank et al., Reference Jank, Barrios, do Valle, Simeão and Alves2014; Ferreira et al., Reference Ferreira, Moraes, Chiari, Simeão, Vigna and de Souza2021; Table 1; Supplementary Material).
A. donax, of Asian origin, is the only C₃ grass species on the list without forage use (Table 1). It is recognized as one of the world’s 100 worst invasive species (Global Invasive Species Database, 2025). The earliest record of the species in South America dates to 1836, in the state of Rio de Janeiro (Supplementary Material). It was recorded in the Bolivian Gran Chaco in 1902 and in the Venezuelan Llanos in 1956. Present in the Cerrado region since 1969 (Supplementary Material), its expansion has accelerated over the past two decades, primarily due to the lack of effective control measures (Calazans et al., Reference Calazans, Lopes, Girotto, de Paula, Franco and Ferreira2023; Jiménez-Ruiz et al., Reference Jiménez-Ruiz, Hardion, Del Monte, Vila and Santín-Montanyá2021). The species poses a serious threat to biodiversity due to its aggressive habitat colonization and displacement of native vegetation (Simões et al., Reference Simões, Hay and Andrade2013; Girotto et al., Reference Girotto, Franco, Nunez, Oliveira, Scheffer de Souza, Fachin-Espinar and Ferreira2021; Jiménez-Ruiz et al., Reference Jiménez-Ruiz, Hardion, Del Monte, Vila and Santín-Montanyá2021).
Ecological impacts of invasive grasses
The invasion of alien perennial grasses in South American savannas has triggered extensive ecological impacts that extend beyond the mere replacement of native vegetation. They disrupt key ecosystem processes by altering energy flows and biogeochemical cycles, with far-reaching consequences for ecosystem functioning over time (Franco et al., Reference Franco, Rossatto, Silva and Ferreira2014; Garcia et al., Reference Garcia, Xavier, Camargo, Vieira and Pivello2022). They often form dense stands and spread quickly in the upper soil layers. Their roots and rhizomes inhibit the recruitment and establishment of native species, leading to reduced structural heterogeneity and ecological complexity above and below ground (Durigan and Ratter, Reference Durigan and Ratter2016; Jiménez-Ruiz et al., Reference Jiménez-Ruiz, Hardion, Del Monte, Vila and Santín-Montanyá2021; Le Stradic et al., Reference Le Stradic, Damasceno, Cancian, Donadieu, Kollmann and Fidelis2025). In the Cerrado, the dominance of U. brizantha and M. minutiflora has been consistently linked to declines in native herbaceous diversity and to shifts in vegetation architecture (Damasceno et al., Reference Damasceno, Souza, Pivello, Gorgone-Barbosa, Giroldo and Fidelis2018; Lopes et al., Reference Lopes, Demarchi, Piedade, Schöngart, Wittmann, Munhoz, Ferreira and Franco2023). These species also alter the composition and dynamics of the soil seed bank. Studies have shown that in areas that are heavily invaded (at least 50% of invasive species cover) by M. minutiflora or U. brizantha, the seed bank becomes dominated by them, with their seeds also being found in significant quantities in adjacent, non-invaded sites (Dairel and Fidelis, Reference Dairel and Fidelis2020). Such patterns suggest that these grasses have a high potential for reinvasion and cause long-term suppression of native flora.
Such changes also affect fauna, which depend on native vegetation structure and composition, reducing functional diversity and leading to simplified ecological networks (Zenni et al., Reference Zenni, Herrera, Dechoum, Ziller, ACL, Núñez, Pauchard, Clements, Upadhyaya, Joshi and Shrestha2022). Landscape homogenization and the loss of structural heterogeneity diminish the resilience of savannas to disturbances, particularly in regions already affected by land-use change (Figure 2). Critical ecosystem services are compromised, including microclimatic regulation, carbon sequestration and soil biogeochemical cycles (Pompeu et al., Reference Pompeu, Assis and Ometto2024).
Figure 2. Diagram summarizing key ecological processes associated with the establishment and spread of invasive grasses in South American savannas, highlighting their effects on fire regimes, nutrient cycling, and landscape structure. Panels A–C show: (A) the presence of M. minutiflora and Urochloa spp following escape from cultivated pastures into natural areas; (B) a fire event intensified by elevated fuel loads in a grass-invaded landscape; and (C) a post-disturbance scenario characterized by simplified vegetation structure, biodiversity loss, and landscape homogenization. Some visual elements in this figure (e.g., plant silhouettes) were generated using an AI image model (ChatGPT, OpenAI) and assembled by the authors.
Perennial grass invasions should therefore be approached from a systemic perspective that recognizes their cumulative effects on the functional integrity of savanna ecosystems. The subsequent sections provide an overview of the principal ecological consequences associated with these invasions, as evidenced in recent scientific research.
Alteration of biogeochemical cycles
Invasive grasses induce substantial changes in soil microbiome, soil organic carbon storage and nutrient dynamics, particularly affecting the cycles of nitrogen and phosphorus (Baptistella et al., Reference Baptistella, Andrade, Favarin and Mazzafera2020; Merloti et al., Reference Merloti, Bossolani, Mendes, Rocha, Rodrigues, Asselta, Crusciol and Tsai2024), which may undermine the competitiveness of native species adapted to the oligotrophic soils typical of tropical savannas. In a pot experiment with different soils, A. donax was capable of altering the chemical properties of autoclaved and bauxite-residue-contaminated soils, increasing soil enzyme activities and the availability of key nutrients such as nitrogen and potassium (Alshaal et al., Reference Alshaal, Domokos-Szabolcsy, Márton, Czakó, Kátai, Balogh, Elhawat, El-Ramady, Gerőcs and Fári2014). The species also enhanced microbial biomass development, suggesting not only a high tolerance to degraded edaphic conditions but also an active role in restructuring biogeochemical processes, thereby promoting its persistence and expansion in disturbed environments.
Similarly, abandoned pastures dominated by invasive species such as U. brizantha, U. decumbens, U. humidicola, A. gayanus, M. minutiflora and H. rufa exhibited higher levels of soil enzyme activity associated with the cycling of carbon, phosphorus and nitrogen, when compared with native savannas (D’Angioli et al., Reference D’Angioli, Giles, Costa, Wolfsdorf, Pecoral, Verona, Piccolo, Sampaio, Schmidt, Rowland, Lambers, Kandeler, Oliveira and Abrahão2022). These altered conditions favor these fast-growing alien grasses with resource-acquisition strategies, while simultaneously undermining the re-establishment of native flora. Restoration efforts in such areas have proven largely ineffective at restoring soil organic matter and microbial biomass to levels observed in undisturbed Cerrado sites, and may even intensify nutrient cycling, thereby creating positive responses that reinforce the dominance of invasive species (D’Angioli et al., Reference D’Angioli, Giles, Costa, Wolfsdorf, Pecoral, Verona, Piccolo, Sampaio, Schmidt, Rowland, Lambers, Kandeler, Oliveira and Abrahão2022).
U. decumbens has been noted for its ability to rapidly transform the ecosystems it invades, altering soil carbon and nitrogen distributions along the soil profile (Garcia et al., Reference Garcia, Xavier, Camargo, Vieira and Pivello2022). M. minutiflora has been associated with shifts in ammonium and nitrate dynamics, as well as increased litter decomposition, pointing to a sustained intensification of nitrogen cycling and reorganization of soil organic matter pools (Sena-Souza et al., Reference Sena-Souza, Rodovalho, Andrade, Pinto and Nardoto2023). Furthermore, field observations and a mesocosm experiment provided evidence that M. minutiflora may benefit from the enhanced soil P and N availability that more diverse Cerrado plant communities provide (Lannes et al., Reference Lannes, Karrer, Teodoro, Bustamante, Edwards and Venterink2020). These processes exemplify the capacity of invasive grasses to reconfigure fundamental biogeochemical pathways.
Fire regime shifts
Natural and anthropogenic fires have historically shaped vegetation structure, floristic composition and biogeochemical cycles in South American savannas. Grasses, which constitute the bulk of the flammable biomass, play a pivotal role in determining fire behavior (Hoffmann et al., Reference Hoffmann, Jaconis, McKinley, Geiger, Gotsch and Franco2012). Fires in this ecosystem are typically fast-moving and of short duration, consuming primarily the ground-layer vegetation. In the Cerrado, flame temperatures can range from 85 °C to over 800 °C, generally recorded at approximately 60 cm above ground level, while subsurface temperatures rarely reached 55 °C at a soil depth of 1 cm (Miranda et al., Reference Miranda, Miranda, Dias and Dias1993). These thermal characteristics, coupled with fire-adaptive traits such as thick bark and underground storage organs, enable many native species to withstand frequent fires and support rapid post-fire regeneration (Eiten Reference Eiten1972; Franco et al., Reference Franco, Rossatto, Silva and Ferreira2014; Pilon et al., Reference Pilon, Cava, Hoffmann, Abreu, Fidelis and Durigan2021).
Invasive grasses such as M. minutiflora and U. brizantha profoundly alter fire dynamics by increasing fuel continuity and biomass, thereby generating fires that are hotter, more intense and more frequent. Such alterations disrupt the natural fire regime and establish positive feedback loops that favor their persistence, while diminishing the survival and recovery potential of native flora not adapted to more intense fire conditions (Rossi et al., Reference Rossi, Martins, Viana, Rodrigues and Figueira2014; Gorgone-Barbosa et al., Reference Gorgone-Barbosa, Pivello, Bautista, Zupo, Rissi and Fidelis2015; Damasceno and Fidelis, Reference Damasceno and Fidelis2020). The shifts in fire ecology, driven by invasive grasses, constitute a major driver of ecological transformation and pose a significant threat to long-term functional resilience of South American savannas (Williams et al., Reference Williams, Watson, Beyer, Grantham, Simmonds, Alvarez, Venter, Strassburg and Runting2022; Pompeu et al., Reference Pompeu, Assis and Ometto2024).
Climate change: synergies with invasive grasses
The intensification of climate change acts as an amplifying driver of ecological pressure that works synergistically with biological invasions to reshape South American savannas. Rising global average temperatures, elevated atmospheric CO₂ concentrations and an increased frequency of extreme climatic events, such as prolonged droughts and heatwaves, are creating environmental conditions that facilitate the establishment and range expansion of drought-tolerant, fire-prone invasive perennial grasses across South American savannas (Brook et al., Reference Brook, Sodhi and Bradshaw2008; Bond & Midgley, Reference Bond and Midgley2012).
These C₄ invasive grasses exhibit highly efficient photosynthetic metabolism under elevated temperatures and high solar radiation, along with notable tolerance to water limitation (Baruch et al., Reference Baruch, Ludlow and Davis1985; Guenni et al., Reference Guenni, Marín and Baruch2002; Beloni et al., Reference Beloni, Santos, Rovadoscki, Balachowski and Volaire2018; Cordeiro et al., Reference Cordeiro, Duarte, Della-Torre, França and França2024). A. donax, although a C₃ species, exhibits photosynthetic rates that are comparable to those of invasive C₄ grasses (Rossa et al., Reference Rossa, Tüffers, Naidoo and Von Willert1998; Haworth et al., Reference Haworth, Marino, Riggi, Avola, Brunetti, Scordia, Testa, Gomes, Loreto, Cosentino and Centritto2019). It is tolerant to both flooding and drought and develops multiple roots capable of penetrating deep into the soil profile (Jiménez-Ruiz et al., Reference Jiménez-Ruiz, Hardion, Del Monte, Vila and Santín-Montanyá2021), enabling access to deeper, more predictable water sources during dry periods. Significant water uptake below 0.5 m depth has been reported for several Urochloa species in greenhouse experiments (Guenni et al., Reference Guenni, Marín and Baruch2002), while field trials have documented root depths of up to 2.8 m for U. decumbens (Rodrigues et al., Reference Rodrigues, Marioti and Roosevelt Júnior2011) and 4.9 m for U. ruziziensis (Silva et al., Reference Silva, de Oliveira, Serafim, Carducci, da Silva, Barbosa, de Melo, dos Santos, Reis, de Oliveira, Guimarães and Castanheira2019).
This combination of acquisitive traits, formation of dense stands, fast spreading and expansion in the upper soil layers, access to deeper water sources confer a competitive advantage to invasive grasses over native species, particularly under projected scenarios of increased aridity and drought stress (Baruch, Reference Baruch, Solbrig, Medina and Silva1996; Damasceno & Fidelis, Reference Damasceno and Fidelis2020). Moreover, alien grasses cultivated extensively for forage are subject to ongoing genetic improvement to enhance their resilience to diverse environmental conditions and to maintain high productivity under changing climatic regimes (Ferreira et al., Reference Ferreira, Moraes, Chiari, Simeão, Vigna and de Souza2021).
Predictive modeling indicates that the potential distribution range of these grasses could expand significantly, reaching ecotonal zones bordering tropical forests and wetland regions of the Amazon (Lopes et al., Reference Lopes, Demarchi, Piedade, Schöngart, Wittmann, Munhoz, Ferreira and Franco2023). From these ecological interfaces, such species may establish novel fire-mediated feedback loops, opening corridors for expansion into previously less susceptible environments.
Experimental studies assessing the effects of elevated CO₂ and warming on these invasive grasses are limited and are difficult to compare due to variations in experimental design. Several investigations have focused on their use as forage or, in the case of A. donax, as a bioenergy crop (Webster et al., Reference Webster, Driever, Kromdijk, McGrath, Leakey, Siebke, Demetriades-Shah, Bonnage, Peloe, Lawson and Long2016). Elevated CO₂ has been shown to enhance germination, seedling size and biomass accumulation in adult plants of H. rufa and M. minutiflora (Baruch and Jackson, Reference Baruch and Jackson2005). Under elevated CO₂ (mean 728 μmol mol⁻1) and higher temperatures (3 °C above ambient), U. brizantha exhibited increased germination and root growth. In contrast, elevated CO₂ did not affect germination or root growth in U. decumbens and M. maximus, although higher temperatures accelerated the germination of U. decumbens (de Faria et al., Reference de Faria, Fernandes and França2015).
Biomass accumulation in U. brizantha and U. decumbens was unaffected by increased CO₂ and/or temperature, while elevated CO₂ stimulated biomass production in M. maximus and improved water use efficiency in all three species (de Faria et al., Reference de Faria, Marabesi, Gaspar and França2018). A 2 °C increase above ambient temperature positively influenced root and shoot dry mass in M. maximus (Carvalho et al., Reference Carvalho, Barreto, Prado, Habermann, Branco and Martinez2020). Additionally, elevated CO₂ (mean 532 μmol mol⁻1) significantly improved water use efficiency in this species by reducing stomatal conductance, with smaller but positive effects on photosynthesis (Habermann et al., Reference Habermann, de Oliveira, Bianconi, Contin, Lemos, Costa, Oliveira, Riul, Bonifácio-Anacleto, Viciedo and Approbato2024).
Regardless of CO₂ concentration, competition with a native C₃ nitrogen-fixing herbaceous legume (Stylosanthes capitata) reduced leaf development and dry matter production in M. minutiflora under both ambient (350 μmol mol⁻1) and elevated (1,000 μmol mol⁻1) CO₂ conditions in a greenhouse experiment (de Oliveira et al., Reference de Oliveira, Rios, Pereira and Souza2021). However, outcomes may differ at the community level, particularly following any event that increases the availability of nutrients to the soil. Both M. minutiflora and U. decumbens have shown strong responses to phosphorus, combined nitrogen and phosphorus and to the addition of macro- and micronutrients in invaded savannas (Lannes et al., Reference Lannes, Bustamante, Edwards and Venterink2016). In field trials, M. minutiflora spread rapidly following fertilization and significantly reduced species richness in the herbaceous layer of savanna vegetation (Bustamante et al., Reference Bustamante, de Brito, Kozovits, Luedemann, de Mello, de Siqueira Pinto, Munhoz and Takahashi2012; Massi et al., Reference Massi, Eugênio, Franco and Hoffmann2021).
The effects of CO₂ enrichment appear to be more pronounced in the C₃ species A. donax. In a closed-top CO₂ growth chamber experiment, increasing CO₂ from approximately 400 to 800 μmol mol⁻1 reduced transpiration rates, delaying drought effects and increasing per-mass water-use efficiency from 6.5 g dry biomass L⁻1 to 12.5 g L⁻1 H₂O (Nackley et al., Reference Nackley, Vogt and Kim2014). CO₂ and nitrogen interact to affect the growth of A. donax. Plants grown under elevated CO₂ (745 μmol mol⁻1) in closed-top growth chambers showed a positive CO₂ enrichment effect on aboveground biomass, while the CO₂ effect for total biomass was significant only under high nitrogen. These plants accumulated approximately 100% more biomass and allocated around 50% more biomass to rhizomes compared to those grown under ambient CO₂ (414 μmol mol⁻1) and low nitrogen (Nackley et al., Reference Nackley, Hough-Snee and Kim2017).
Elevated CO₂ levels, increased nitrogen deposition, more frequent fires, rising aridity and higher temperatures are all expected in the near future. Although elevated CO₂ concentrations may benefit a few fast-growing tree species under specific conditions, they are also likely to promote the proliferation of invasive grasses in open ecosystems, shifting the competitive balance and reducing the regenerative capacity of native vegetation (Bond and Midgley, Reference Bond and Midgley2012). The experimental studies mentioned above show that invasive grasses increase their productivity in CO₂-enriched environments, especially when nutrients are not limited. In this context, increasing nitrogen deposition from anthropogenic sources may further enhance the growth and spread of invasive grasses (Eller and Oliveira, Reference Eller and Oliveira2017; Ferreira et al., Reference Ferreira, Faria, Vasconcelos, Bruna, Costa and Moreira2024), especially A. donax. The high water-use efficiency of invasive C₄ grasses may also support their establishment and competitive advantage under a future scenario of drier and warmer conditions, elevated CO₂ and more frequent fires.
It is important to stress that invasive grasses thrive and rapidly spread in disturbed sites with higher nutrient levels due to fertilization from previous management for pasture or crops (Vanlauwe et al., Reference Vanlauwe, Aihou, Houngnandam, Diels, Sanginga and Merckx2001, Reference Vanlauwe, Diels, Lyasse, Aihou, Iwuafor, Sanginga, Merckx and Deckers2002; Lira-Martins et al., Reference Lira-Martins, Xavier, Mazzochini, Verona, Andreuccetti, Martins, de Barros, Furtado, Stein, Abrahão, Sampaio, Schmidt, Rowland and Oliveira2025), to soil nutrient enrichment after a fire (Pellegrini and Jackson, Reference Pellegrini, Jackson, Dumbrell, Turner and Fayle2020; Giles et al., Reference Giles, Silva, Mazzochini, Flores, Rowland, Costa, Cure, Monge, Schmidt, Abrahão, Sampaio, Côrtes and Oliveira2025), nutrient-richer soils that underlie many forests undergoing degradation (Veldman and Putz, Reference Veldman and Putz2011) or runoff from nearby agricultural fields. Thus, the interactive effects of climate change, land-use transformation and biological invasions collectively drive landscape homogenization, reduce functional diversity and destabilize ecosystem processes (Figures 2 and 3). Investigating different trophic levels is therefore crucial for understanding how invasive grasses disrupt ecosystem functioning, as well as for developing effective strategies to mitigate their impacts.
Figure 3. Structural contrast in the Brazilian Cerrado. (A) Native savanna vegetation; (B) area invaded by the alien grasses A. donax and M. maximus, showing dense cover by these two grasses and reduced structural heterogeneity of the vegetation.
Challenges and strategies for controlling invasive grasses
One of the major challenges in the restoration of South American savannas lies in the revegetation of abandoned pastures and crop fields that, for various reasons, are considered suitable for ecological restoration (Williams et al., Reference Williams, Watson, Beyer, Grantham, Simmonds, Alvarez, Venter, Strassburg and Runting2022; Silva et al., Reference Silva, Rodrigues, Bringel, Sampaio, Sano and Vieira2023). Here, we focus on the Cerrado, where most studies in restoration of native savanna vegetation were carried out in recent years (Medeiros et al., Reference Medeiros, Ordóñez-Parra, Buisson and Silveira2024). The presence of extensive areas at varying levels of degradation and invasion by alien perennial grasses is a common feature in many PAs in the Cerrado. Depending on the management policies of each PA, these areas are subjected either to passive restoration or to targeted interventions aimed at re-establishing the original savanna vegetation cover. Various methods have been employed to promote the recovery of native vegetation, with direct seeding being the most widely adopted. However, competition with invasive grasses significantly limits the growth and re-establishment of native flora (Passaretti et al., Reference Passaretti, Pilon and Durigan2020; Cava et al., Reference Cava, Pilon, Priante, Ribeiro and Durigan2020).
Recolonization by native species alone tends to be insufficient to prevent the reinvasion of alien perennial grasses (Wiederhecker et al., Reference Wiederhecker, Cardoso Ferreira, Barbosa Rodrigues, Bonesso Sampaio, Belloni Schmidt, Ribeiro, Ogata, Rodrigues, Silva-Coelho, Sousa Abreu, Montenegro and Mascia Vieira2024; Cianciaruso et al., Reference Cianciaruso, Vellosa and Coutinho2025). It must be accompanied by strategies that control the emergence and establishment of invasive grasses without compromising native species recovery. Compared to invasive African grasses, native Cerrado grasses produce fewer viable seeds and have lower seedling emergence rates (Aires et al., Reference Aires, Sato and Miranda2014; Fontenele et al., Reference Fontenele, Figueirôa, Pereira, Nascimento, Musso and Miranda2020), which suggests an initial competitive disadvantage in outcompeting invasive species (Martins et al., Reference Martins, Hay and Carmona2009; Aires et al., Reference Aires, Sato and Miranda2014; Dantas-Junior et al., Reference Dantas-Junior, Musso and Miranda2018; Musso et al., Reference Musso, de Macedo, Almeida, Rodrigues, Camargo, Pôrto and Miranda2019). Some invasive species, such as U. decumbens, can also flower multiple times throughout the year (Dantas-Junior et al., Reference Dantas-Junior, Musso and Miranda2018; Xavier et al., Reference Xavier, Leite and da Silva Matos2019).
The removal of a single invasive grass species was ineffective in restoring native savanna diversity when other invasive species co-occurred. In a 2-year field experiment in the Cerrado, M. minutiflora and U. decumbens replaced each other in terms of ground cover when only one species was removed, indicating compensatory dynamics among dominant invasive grasses (Zenni et al., Reference Zenni, da Cunha, Musso, de Souza, Nardoto and Miranda2020).
Straw mulching increased soil moisture and strongly reduced the emergence of U. decumbens. However, it inhibited the emergence of tree species with flat seeds and phanerocotylar seedlings (Silva and Vieira, Reference Silva and Vieira2017). Mulching may have a greater impact on the seedling emergence of native Cerrado grasses than on invasive grasses. For example, seedling emergence in nine Cerrado grass species declined with sowing depth, and only two species emerged from a depth of 3 cm (Fontenele et al., Reference Fontenele, Figueirôa, Pereira, Nascimento, Musso and Miranda2020). In contrast, the seedling emergence of U. decumbens was unaffected by sowing depths of 0, 1, 2 or 3 cm (Dantas-Junior et al., Reference Dantas-Junior, Musso and Miranda2018), and only slightly affected in A. gayanus at a depth of 3 cm (Musso et al., Reference Musso, de Macedo, Almeida, Rodrigues, Camargo, Pôrto and Miranda2019). Despite reductions in emergence rates, U. decumbens (Marques et al., Reference Marques, Souza, Pereira, Marchi, Martins and Martins2022) and U. ruziziensis (Marques et al., Reference Marques, Gomes, Martins, de Marchi and Martins2024) were able to germinate from seeds buried at depths of up to 12 cm. Although M. minutiflora showed a more pronounced decline in seedling emergence with increasing depth, germination was still recorded in seeds buried at 3 cm (Martins et al., Reference Martins, Hay and Carmona2009).
Dodonov et al. (Reference Dodonov, Braga, Sales, Xavier and Matos2020) conducted a field study on the regeneration of the savanna woody layer in an area previously occupied by a Eucalyptus grandis plantation. Although the plantation had been removed, scattered E. grandis trees and a dense cover of U. decumbens and M. minutiflora remained. The authors concluded that shading from the remaining eucalyptus trees might be more detrimental to the invasive grasses than to the regeneration of native woody savanna species; however, the study did not assess the impact on the herbaceous layer.
In degraded Cerrado grasslands, soil chemistry manipulation through acidification with iron sulphate (Chhabra, Reference Chhabra and Chhabra2021) has been effective in reducing the biomass of nutrient-demanding invasive grasses, without negatively affecting native plant biomass. This approach restores the original acidic, nutrient-poor conditions typical of Cerrado soils, thereby favoring native species (Lira-Martins et al., Reference Lira-Martins, Xavier, Mazzochini, Verona, Andreuccetti, Martins, de Barros, Furtado, Stein, Abrahão, Sampaio, Schmidt, Rowland and Oliveira2025).
Mechanical mowing is generally ineffective for controlling invasive species capable of resprouting, such as U. decumbens, U. brizantha and A. donax. These species resprout vigorously from rhizomes after being cut back to soil level. In fact, mowing can intensify the invasion of A. donax, which is able to regenerate from stem fragments left on the ground, potentially leading to rapid spread in wetland areas. Stem fragments can remain viable for up to 30 days underwater (Calazans et al., Reference Calazans, Lopes, Girotto, de Paula, Franco and Ferreira2023), while rhizome fragments can survive submerged for up to 16 weeks and successfully produce new plants (Mann et al., Reference Mann, Barney, Kyser and Di Tomaso2013).
Ploughing to uproot rhizomes of invasive perennial grasses causes extensive disturbance to both invasive and native vegetation and has proven largely ineffective, as invasive grasses can still regenerate from the seed bank (Mazzochini et al., Reference Mazzochini, Lira-Martins, de Barros, Oliveira, Xavier, Furtado, Verona, Viani, Rowland and Oliveira2024). In contrast, hoeing – manual removal of invasive grass clumps – although labor-intensive, was successful in restoring both ground cover and native species richness without the need for planting (Assis et al., Reference Assis, Pilon, Siqueira and Durigan2021).
Prescribed fire is not effective in controlling most invasive grasses (Pivello et al., Reference Pivello, Shida and Meirelles1999; Martins et al., Reference Martins, Hay, Scaléa and Malaquias2017; Assis et al., Reference Assis, Pilon, Siqueira and Durigan2021; Giles et al., Reference Giles, Silva, Mazzochini, Flores, Rowland, Costa, Cure, Monge, Schmidt, Abrahão, Sampaio, Côrtes and Oliveira2025). However post-fire recovery dynamics can vary among species in invaded areas. M. minutiflora tends to decline in abundance following fire events (Damasceno and Fidelis, Reference Damasceno and Fidelis2020). Areas dominated by M. minutiflora often become more open post-fire, enabling native species to colonize the newly available spaces and thereby increasing overall community abundance and diversity (Damasceno and Fidelis, Reference Damasceno and Fidelis2020). This trend is consistent with findings from a field experiment conducted in southeastern Brazil that examined the influence of the fire season on native savanna communities invaded by M. minutiflora or U. brizantha (Dezotti et al., Reference Dezotti, Fidelis, Damasceno and Siqueira2024). Over time, plots invaded by M. minutiflora gained more species than those dominated by U. brizantha, with the highest species gains observed in areas subjected to late-dry season fires. However, fire events can facilitate the reestablishment of invasive grasses in savanna areas undergoing restoration, including M. minutiflora (Assis et al., Reference Assis, Pilon, Siqueira and Durigan2021; Giles et al., Reference Giles, Silva, Mazzochini, Flores, Rowland, Costa, Cure, Monge, Schmidt, Abrahão, Sampaio, Côrtes and Oliveira2025).
The use of grass-selective herbicides, whether applied alone or in combination with other techniques, has been effective in controlling invasive grasses and promoting the regeneration of the woody stratum in savannas (Durigan et al., Reference Durigan, Contieri, GADC and MAO1998; Pereira et al., Reference Pereira, Mucida, de Oliveira, Barroso, Santana, Titon and dos Santos2025). Herbicides may have to be repeatedly applied to provide effective long-term control and could have detrimental non-target effects on the regeneration of native ground-layer species (Assis et al., Reference Assis, Pilon, Siqueira and Durigan2021), which would require follow-up interventions, such as the active reintroduction of native Cerrado grasses, which represent an important component of the ground-layer vegetation. Moreover, the use of herbicides may be restricted as a weed-control measure in PAs because of the high risk of environmental contamination (Gandhi et al., Reference Gandhi, Khan, Patrikar, Markad, Kumar, Choudhari, Sagar and Indurkar2021).
Herbicide drift, the unintended dispersion of chemicals beyond the target site, also poses a risk to adjacent plant communities (Boutin et al., Reference Boutin, Strandberg, Carpenter, Mathiassen and Thomas2013; Lugar et al., Reference Lugar, Nelson and Wagner2023). Long-term monitoring of herbicide use for controlling invasive grasses is crucial, both in PAs and in disturbed sites undergoing restoration. In addition to assessing vegetation changes following herbicide application, further studies should investigate potential non-target effects, herbicide persistence in soil and water, degradation and transformation pathways, and how they affect soil microbiota and ecosystem processes.
Thus, to mitigate the impacts of invasive grasses, integrated strategies are required, including:
-
1. Proactively reintroducing native species as functional groups, including different growth forms, in areas that have been degraded to improve ecosystem resilience and aid in restoring the structure and diversity of native vegetation (Silveira et al., Reference Silveira, Arruda, Bond, Durigan, Fidelis, Kirkman, Oliveira, Overbeck, Sansevero, Siebert, Siebert, Young and Buisson2020; Pilon et al., Reference Pilon, Campos, Durigan, Cava, Rowland, Schmidt, Sampaio and Oliveira2023).
-
2. Monitoring and public policies aimed at preventing the spread of these invasive grasses, particularly within PAs and ecologically sensitive zones (Strassburg et al., Reference Strassburg, Brooks, Feltran-Barbieri, Iribarrem, Crouzeilles, Loyola, Latawiec, Oliveira Filho, Scaramuzza, Scarano, Soares-Filho and Balmford2017; Rossiter-Rachor et al., Reference Rossiter-Rachor, Adams, Canham, Dixon, Cameron and Setterfield2023). The use of multispectral imagery obtained via drones is a promising approach for detecting the presence and spatial patterns of invasive grasses in grasslands and savanna ecosystems (Rezende de Ataíde, Reference Rezende de Ataíde, Rodrigues, Silva, Coelho, Wiederhecker and Vieira2024).
-
3. Developing trait-based climate niche models (Medeiros et al., Reference Medeiros, Henry, Trueba, Anghel, Guerrero, Pivovaroff, Fletcher, John, Lutz, Méndez Alonzo and Sack2023) to anticipate the expansion of invasive grasses and persistence of savanna plant assemblages under future climate change scenarios, thereby enabling the design of effective containment and management strategies in future climates.
Conclusion
Despite a growing body of scientific evidence on the ecological impacts of invasive grasses in South American savannas, progress in their control and management remains limited by conceptual barriers, institutional fragmentation and the lack of public policies specifically tailored to open ecosystems. The persistence and resilience of these grasses, even within protected and monitored areas, underscore the limitations of conventional conservation approaches and highlight the urgent need to revise current environmental management models.
Successful management to contain invasive grasses during savanna restoration requires addressing both the species-rich groundlayer and the open canopy of scattered trees and shrubs. Although restoration of the ground-layer vegetation remains a major ecological challenge, we now have enough information to successfully germinate a functionally diverse and species-rich assemblage of ground-layer species (Kozovits et al., Reference Kozovits, Figueiredo and Messias2026), a crucial step for restoring biodiversity and managing invasive grasses. However, complete eradication from protected or recovering areas is unlikely. The naturalization of these species in the Cerrado, and likely in other South American savannas, appears to be an almost irreversible process. A more realistic scenario involves containing their spread, particularly in regions of high conservation value, ecological restoration zones and areas managed using prescribed fire.
Addressing this challenge requires going beyond isolated and reactive actions towards integrated management strategies grounded in ecological knowledge and supported by interinstitutional cooperation and active societal participation. Priority should be given to the active restoration of native species, the application of predictive ecological modeling, systematic long-term monitoring and the implementation of more effective regulatory mechanisms. Furthermore, the ecological, climatic and cultural significance of South American savannas must be more firmly integrated into conservation agendas and land-use planning frameworks.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/dry.2026.10024.
Data availability statement
The authors confirm that the data supporting the findings of this study are available within the article and its supplementary material.
Acknowledgements
We thank the reviewers for careful reading of our manuscript and their insightful comments and suggestions
Author contribution
Both authors contributed equally to this work.
Financial support
Funding for this research was provided by the National Council for Scientific and Technological Development (CNPq, Brazil) grant numbers 312336/2023–3, 444726/2024–1. Fundação de Apoio à Pesquisa do Distrito Federal grant number 00193–00001823/2023–09.
Competing interests
The authors declare none.
Open access
Comments
Dear Professor David Eldridge/Professor Osvaldo Sala
We are pleased to submit our review article entitled “On the Rise: The Impact of Exotic Grasses on Neotropical Savannas” for consideration for publication in Drylands.
In this manuscript, we provide a comprehensive synthesis of the ecological impacts of invasive exotic grasses in neotropical savannas, with particular emphasis on the Brazilian Cerrado. Our review integrates current findings from ecological, physiological, and modelling studies to examine how these species alter fundamental ecosystem processes, including fire regimes, nutrient cycling, and native plant community dynamics. We also discuss the implications of climate change for invasion dynamics and ecosystem resilience.
This article offers a timely and critical perspective on the growing threat posed by invasive grasses to tropical savannas, ecosystems that remain underrepresented in conservation agendas despite their high biodiversity and functional importance. We believe that the review will be of interest to a broad audience of ecologists, conservation scientists, land managers, and policy-makers working on biological invasions, ecological restoration, and dryland ecosystem management.
Thank you for considering our submission.
Sincerely,
Cristiane Silva Ferreira
Corresponding author